AI Dictation v2 is here: turn speech into text privately, on your own device
A speech-to-text app built on the cross-platform AI Suite v2 stack: dictate from your microphone, capture your computer's audio, or transcribe a recording — and get clean, editable text with subtitles, playback and an encrypted local history, all on your device.
Today we're releasing AI Dictation v2 — a speech-to-text app built on the cross-platform AI Suite v2 stack. Speak and watch the words appear, capture a meeting your computer is playing, or drop in a recording — and get clean, editable text. The point is simple: turn talk into text without sending your voice to the cloud.
What it does
AI Dictation transcribes from three sources: your microphone (dictate live and watch the text stream in), your computer's own audio (a meeting, a video, a podcast playing on your PC), or an audio file you already have. With the free on-device engine, the whole transcription runs on your machine — turn the network off and it keeps working.
The first time you record, the app picks and downloads a capable on-device speech model for you, then captions your speech as you talk. Choose the spoken language or let it auto-detect, turn on timestamps when you need them, and save the result as plain text, as SubRip (.srt) or WebVTT (.vtt) subtitles to caption a video, or as JSON with every segment's timing for your own tools.
New in v2
- Three ways to capture. Microphone, your computer's system audio, or a file — switch sources in a click, with a live meter that shows your input is being heard.
- Replay and re-transcribe. Every recording is saved, so you can play it back with a waveform scrubber and watch each word highlight as it plays. Re-run a recording through a more accurate model any time — the new result is saved as a separate entry, so the original is never overwritten.
- Subtitles and clean exports. Export SubRip and WebVTT subtitles with accurate cue timing, JSON for your tooling, or plain text. A cleanup control decides how non-speech sounds like [Music] or (laughs) are handled, with the raw model output one panel away.
- History you control. Every transcript is saved locally and encrypted at rest, ready to reopen, copy or continue. Search any transcript with find — which also filters your recent list to recordings that mention what you typed.
- Your choice of engine. Start free with an on-device Whisper model. Pro unlocks larger models, the OpenAI cloud engine, and a GPU-capable local engine for the toughest audio — with built-in GPU safety limits and a keep-models-loaded option so repeat jobs start instantly.
- In 16 languages, and scriptable. The whole interface follows your operating system's language and transcribes dozens of spoken languages. A headless command-line tool transcribes from the terminal and doubles as an MCP server so AI agents can call it on your own infrastructure.
Every feature is free
As with the rest of the suite, no feature is locked behind Pro. Live dictation, system-audio and file transcription, subtitle export, playback, re-transcribe, encrypted history, find and the command-line tool are all in the free app, with a capable on-device model and no time limit or nag screens. Pro is simply about choice of engine — larger local models, the OpenAI cloud engine, and a GPU-accelerated local engine when you want extra accuracy. You never lose a capability by staying free.
Why on-device matters
Interviews, meetings, medical notes, legal dictation — the audio people most want transcribed is often the audio that can't go to the cloud. AI Dictation brings the AI to your recordings instead: nothing leaves the device with the on-device engine, your transcripts and recordings are encrypted at rest, and there are no per-minute transcription fees. It's the same on-premises, privacy-first principle behind everything Software Tailor builds [1].
Get it
AI Dictation v2 is available now on the Microsoft Store for Windows as a free download, with Pro available when you want premium and cloud engines [2]. Editions for macOS, Linux, browser and mobile are on the roadmap. Read the full feature tour on the product page.
References
- Software Tailor. "AI Dictation — product page." softwaretailor.com/ai-dictation.htm. Accessed 2026-06-22.
- Microsoft. "AI Dictation on the Microsoft Store." apps.microsoft.com/detail/9N4PBW55J3ZL. Accessed 2026-06-22.
Related articles
Turn talk into text, privately.
Free to start, on-device by design. Get it on Windows from the Microsoft Store, or explore the full feature tour first.